Radial artery access and outcomes for left main stem PCI: an analysis of 19,482 cases from the British Cardiovascular Intervention Society national database

Kinnaird, Tim; Anderson, Richard; Sirker, Alexander; Ludman, Peter; de Belder, Mark; Oldroyd, Keith; Curzen, Nick; Banning, Adrian; Mamas, Mamas

Abstract

Background
Although the predictors and outcomes of radial versus femoral access are well defined for unselected PCI procedures and certain sub-groups, there are few data for patients undergoing left main stem percutaneous coronary intervention (LMS-PCI). The aim of the present study was to address these questions by using the BCIS National PCI Audit data for 2007 to 2014 to explore temporal changes in national arterial access site practice in patients undergoing LMS-PCI, to define in contemporary study years the predictors of access site, and to report procedural and clinical outcomes by access site.

Methods
Data were analysed from 19,482LMS-PCI procedures performed in England and Wales between 2007 and 2014. Using the BCIS dataset, patients were categorised into either radial or femoral access during LMS-PCI. Study definitions were used as in the BCIS-NICOR database. To investigate if there was a trend over time of radial access adoption in different sub-groups, linear regression modelling and testing for the slope of the regression line was performed. Multivariate logistic regression was used to identify predictors of access site choice and its association with outcomes. To adjust for missing data we used multiple imputations with chained equations. Individual logistic regressions were done on the imputed data set for each of the MACE events according to the access site to quantify the independent association between access site and outcomes. In order to adjust for baseline characteristics that might influence outcomes and thus to attempt to quantify the independent effects of access site on outcomes, we performed a propensity score analysis using the inverse probability of treatment weights.

Results
The frequency of femoral artery use fell from 78.9% in 2007 to 39.6% in 2014 (p<0.001 for trend) with the adoption of radial access being less rapid in patients with renal failure or a prior history of CABG, and in patients undergoing rotational atherectomy or PCI for CTO disease.Femoral access was also associated with certain characteristics representing PCI complexity including CTO intervention, use of rotational atherectomy and the requirement for circulatory support. However, radial access was associated with a greater number vessels treated and stents used, more unprotected LMS-PCI, and more intervention involving the left main stem plus the proximal LAD, the left main stem plus the proximal LAD/circumflex, and the left main stem plus any other vessel. There was a negatively skewed distribution of operator volume of LMS-PCI cases with a significant number of low volume operators, and a median total operator volume of 9 cases over the three years examined. There was also negatively skewed distribution of centre volume of LMS-PCI cases with a significant number of low volume centres, and a median total centre volume of 51 cases over the three years examined. There was a correlation between tertiles of operator volume and the likelihood of femoral access, and between tertiles of operator volume and femoral access (p<0.001 for both comparisons). There was also a correlation between individual operator volume and femoral access (p=0.005 for trend) and a trend for a correlation between individual centre volume and femoral access (p=0.076). In the most contemporary study years (2012-14) using multivariate analyses, the predictors of femoral artery use for LMS-PCI were age/year (OR 1.01, 95% confidence interval [1.004-1.02], p<0.001), previous CABG (OR 2.44 [2.08-3.83], p<0.001), the need for mechanical LV support (OR 2.83 [2.10-2.60], p<0.001), history of renal disease (OR 2.10 [1.67-2.64], p<0.001), CTO intervention (OR 1.47 [1.19-1.82], p<0.001) and female sex (OR 1.43 [1.27-1.61], p<0.001). Radial artery use was associated with use of intravascular ultrasound (OR 0.55, [0.37-0.81], p=0.002). Operator and centre volumes were not independently associated with femoral access. Length of stay was shorter (2.6±9.2 vs. 3.6±9.0, p<0.001) and same day discharge greater (43.0% vs. 26.6%, p<0.001) in the radial cohort. After propensity score analysis to adjust for differences in baseline demographics, vascular access complications (OR 2.93, [1.75-4.93], p<0.001), vascular access surgery (OR 20.9, [2.98-147.33], p<0.001), in-hospital major bleeding (OR 2.2, [1.35-3.61], p=0.001) and mortality at all study time-points (12-months OR 1.23, [1.05-1.47], p=0.013) more frequent in the femoral cohort. The unadjusted and adjusted Kaplan Meier curves for 12-month survival by access site were significantly different.

Conclusion
In contemporary practice in the United Kingdom, the radial artery is the predominant access site for LMS-PCI. Radial access use is associated with shorter length of stay, less vascular complications, major bleeding and transfusion, and better survival at 12-months than femoral access.

Citation

Kinnaird, T., Anderson, R., Sirker, A., Ludman, P., de Belder, M., Oldroyd, K., Curzen, N., Banning, A., & Mamas, M. (2018, September). Radial artery access and outcomes for left main stem PCI: an analysis of 19,482 cases from the British Cardiovascular Intervention Society national database. Presented at 13th Annual Symposium Transcatheter Cardiovascular Therapeutics (TCT)

• Presentation Conference Type: Presentation / Talk
• Conference Name: 13th Annual Symposium Transcatheter Cardiovascular Therapeutics (TCT)
• Start Date: Sep 21, 2018
• End Date: Sep 25, 2018
• Deposit Date: Jun 26, 2023
• Publisher: Elsevier
• Public URL: https://keele-repository.worktribe.com/output/507683